Systems and methods for predicting collision probabilities associated with roadway intersections

This application claims priority to and the benefit of U.S. Patent Application Ser. No. 63/542,412, filed on Oct. 4, 2023, and U.S. Patent Application Ser. No. 63/609,411, filed on Dec. 13, 2023, each of which is hereby incorporated by reference in its entirety.

The present disclosure generally relates to infrastructure safety. More specifically, the present disclosure relates to determining risk associated with roadway intersections using telematics data.

Telematics data obtained from vehicles may include information, parameters, attributes, characteristics, and/or features associated with the vehicles. For example, telematics data may include data relating to various components of the vehicle (e.g., airbag status, engine data, brake data, transmission data, and the like), location data (e.g., a GPS location), vehicle identifying information, etc.

While telematics data may be used to gain insights about various aspects of the vehicles from which it is collected, the telematics data may also be used to provide insights about the area or areas within which the vehicles operate. For example, using telematics data such as geospatial data, airbag data, and/or the like, a user may be able to determine whether a vehicle has been in a traffic accident, collision, or crash, the location of such an incident, etc.

As will be appreciated, traffic accidents, collisions, or crashes involving vehicles may have serious consequences. For example, crashes may result in serious injuries or, in some cases, fatal injuries to vehicle operators, vehicle passengers, cyclists, and/or pedestrians. As well, collisions may have significant costs associated therewith, such as those associated with vehicle damage, property damage (e.g., damage to cargo carried on the vehicle), insurance ramifications, incurred legal fees, medical treatments, operational delays, internal procedures (e.g., updating of internal SOPs), etc. It is desirable for a number of reasons to avoid traffic accidents, collisions, and crashers wherever possible.

Notably, in the year 2020 in Canada, according to the National Collision Database, 41% of serious injuries occurred at roadway intersections. Thus, roadway intersections represent a significant source of traffic accidents, collisions, or crashes and, as a result, it may be desirable to assess the risk associated with individual roadway intersections such that a vehicle operator or, for example, a fleet manager, may plan a route that minimizes traversal of higher-risk, or unsafe, roadway intersections. However, conventional techniques for assessing roadway safety may generally focus on the risk associated with the intersecting roadways rather than a roadway intersection itself. Such conventional techniques may therefore provide inaccurate risk assessments of roadway intersections, which, as will be appreciated, may cause a vehicle operator, fleet manager, etc. to unnecessarily plan a route through a particularly hazardous roadway intersection.

A need therefore exists for improved systems and methods for determining risk associated with roadway intersections.

In one aspect, the present disclosure relates to a system for predicting collision risk associated with a roadway intersection, the system comprising: at least one data storage operable to store map data and telematics data originating from telematics devices installed in a plurality of vehicles; and at least one processor, in communication with the at least one data storage, operable to: identify, using the map data, one or more roadway intersections; determine, using the telematics data and/or the map data, for each roadway intersection, one or more roadway intersection metrics thereof; determine a hazard rating for each roadway of each roadway intersection; and generate a collision probability for each roadway intersection by inputting into a machine learning model the one or more roadway intersection metrics and the hazard rating of each roadway thereof, the collision probability representing a risk of collision for a vehicle traversing the roadway intersection.

In some embodiments, the one or more roadway intersection metrics comprise an intersection turn percentage, an intersection traversal time, an intersection complexity metric, or a combination thereof.

In some embodiments, the intersection turn percentage comprises a left turn percentage, a right turn percentage, a straight-through percentage, or a combination thereof.

In some embodiments, the intersection traversal time comprises a left turn time, a right turn time, a straight-through time, or a combination thereof.

In some embodiments, the intersection complexity metric comprises an intersection vehicle entropy, an intersection turn entropy, an intersection road entropy, or a combination thereof.

In some embodiments, the at least one processor is further operable to determine, using the telematics data and/or the map data, one or more roadway metrics of each roadway of each roadway intersection.

In some embodiments, the one or more roadway metrics comprise a static roadway metric, a driving behavior metric, a traffic volume metric, a traffic speed metric, a travel time metric, an environmental metric, a congestion metric, a vehicle complexity metric, or a combination thereof.

In some embodiments, the at least one processor is operable to determine the collision probability for each roadway intersection by inputting into the machine learning model the one or more roadway intersection metrics, the hazard rating, and the one or more roadway metrics of each roadway thereof.

In some embodiments, the at least one processor is operable to determine the hazard rating of each roadway of each roadway intersection based at least in part on a number of collisions that have occurred therealong within a selected time period.

In some embodiments, the hazard rating comprises a binary rating.

In some embodiments, the machine learning model comprises a classification model.

In some embodiments, the classification model comprises a random forest (RF) model, a logistic regression (LR) model, a gradient boosting (GB) model, or a combination thereof.

In another aspect, the present disclosure relates to a system for predicting collision risk associated with a roadway intersection, the system comprising: at least one data storage operable to store map data and telematics data originating from telematics devices installed in a plurality of vehicles; and at least one processor, in communication with the at least one data storage, operable to: train a machine learning model to generate a collision probability for a roadway intersection by inputting into the machine learning model training data associated with one or more previously-identified roadway intersections, the training data comprising: one or more previously-determined roadway intersection metrics associated with each previously-identified roadway intersection, a previously-determined hazard rating of each roadway of each previously-identified roadway intersection, and a previously-determined collision probability associated with each previously-identified roadway intersection; identify, using the map data, one or more additional roadway intersections; determine, using the telematics data and/or the map data, for each roadway of each additional roadway intersection, one or more roadway intersection metrics thereof; determine a hazard rating for each roadway of each additional roadway intersection; generate a collision probability of each additional roadway intersection by inputting into the machine learning model the one or more roadway intersection metrics and the hazard rating of each roadway thereof; and modify the training data to include the one or more roadway intersection metrics of each roadway of each additional roadway intersection, the hazard rating of each roadway of each additional roadway intersection, and the collision probability of each additional roadway intersection.

In another aspect, the present disclosure relates to a method for predicting collision risk associated with a roadway intersection, the method comprising operating at least one processor to: receive map data and telematics data originating from telematics devices installed in a plurality of vehicles; identify, using the map data, one or more roadway intersections; determine, using the telematics data and/or map data, for each of the one or more roadway intersections, one or more roadway intersection metrics thereof; determine a hazard rating for each roadway of each roadway intersection; and generate a collision probability for each roadway intersection by inputting into a machine learning model the one or more roadway intersection metrics and the hazard rating of each roadway thereof, the collision probability representing a risk of collision for a vehicle traversing the roadway intersection.

In some embodiments, the one or more roadway intersection metrics comprise an intersection turn percentage, an intersection traversal time, an intersection complexity metric, or a combination thereof.

In some embodiments, the intersection turn percentage comprises a left turn percentage, a right turn percentage, a straight-through percentage, or a combination thereof.

In some embodiments, the intersection traversal time comprises a left turn time, a right turn time, a straight-through time, or a combination thereof.

In some embodiments, the intersection complexity metric comprises an intersection vehicle entropy, an intersection turn entropy, an intersection road entropy, or a combination thereof.

In some embodiments, the method further comprises operating the at least one processor to determine, using the telematics data and/or the map data, one or more roadway metrics of each roadway of each roadway intersection.

In some embodiments, the one or more roadway metrics comprise a static roadway metric, a driving behavior metric, a traffic volume metric, a traffic speed metric, a travel time metric, an environmental metric, a congestion metric, a vehicle complexity metric, or a combination thereof.

In some embodiments, the generating of the collision probability of each roadway intersection comprises operating the at least one processor to input into the machine learning model the one or more roadway intersection metrics, the hazard rating, and the one or more roadway metrics of each roadway thereof.

In some embodiments, the determining of the hazard rating is based at least in part on a number of traffic collisions that have occurred along each roadway of each roadway intersection within a selected time period.

In some embodiments, the hazard rating comprises a binary rating.

In some embodiments, the machine learning model comprises a classification model.

In some embodiments, the classification model comprises a random forest (RF) model, a logistic regression (LR) model, a gradient boosting (GB) model, or a combination thereof.

In another aspect, the present disclosure relates to a method for predicting collision risk associated with a roadway intersection, the method comprising operating at least one processor to: receive map data and telematics data originating from telematics devices installed in a plurality of vehicles; train a machine learning model to generate a collision probability for a roadway intersection by inputting into the machine learning model training data associated with one or more previously-identified roadway intersections, the training data comprising: one or more previously-determined roadway intersection metrics associated with each previously-identified roadway intersection, a previously-determined hazard rating of each roadway of each previously-identified roadway intersection, and a previously-determined collision probability associated with each previously-identified roadway intersection; identify, using the map data, one or more additional roadway intersections; determine, using the telematics data and/or the map data, for each roadway of each additional roadway intersection, one or more roadway intersection metrics thereof; determine a hazard rating for each roadway of each additional roadway intersection; generate a collision probability of each additional roadway intersection by inputting into the machine learning model the one or more roadway intersection metrics and the hazard rating of each roadway thereof; and modify the training data to include the one or more roadway intersection metrics of each roadway of each additional roadway intersection, the hazard rating of each roadway of each additional roadway intersection, and the collision probability of each additional roadway intersection.

In another aspect, the present disclosure relates to a non-transitory computer readable medium having instructions stored thereon executable by at least one processor to implement the methods described herein.

Other aspects and features of the systems and methods of the present disclosure will become apparent to those ordinarily skilled in the art upon review of the following description of specific embodiments.

Traffic accidents, collisions, or crashes involving vehicles may have serious consequences. For example, crashes may result in serious injuries or, in some cases, fatal injuries to a variety of stakeholders, including vehicle operators, vehicle passengers, cyclists, and pedestrians. In addition to bodily harm, collisions may have significant costs associated therewith, such as those associated with vehicle damage, property damage (e.g., damage to cargo carried on the vehicle), insurance ramifications, incurred legal fees, medical treatments, operational delays, internal procedures (e.g., updating of internal SOPs), etc. It is therefore desirable to avoid traffic accidents, collisions, and crashers wherever possible.

One way for a vehicle operator to avoid potential traffic accidents, collisions, or crashes is to avoid infrastructure (e.g., roadways, or portions of roadways) that may be particularly risky to traverse. For example, operators of heavy-duty trucks (e.g., class 6 vehicles) may choose to avoid roadways that are particularly busy (e.g., have a relative high traffic volume), that include sharp or difficult to navigate turns, that are not well-maintained, or any other conditions could lead to a collision.

One type of infrastructure that may be particularly risky to traverse is a roadway intersection. In fact, in the year 2020 in Canada, according to the National Collision Database 41% of serious injuries occurred at roadway intersections. However, as will be appreciated, it may often be difficult, if not impossible, to avoid roadway intersections altogether. It may therefore be desirable to assess the risk associated with individual roadway intersections such that a vehicle operator or, for example, a fleet manager, may plan a route that minimizes traversal of roadway intersections that are particularly unsafe or risky.

Conventional techniques for assessing roadway safety, however, may generally focus on the risk associated with the intersecting roadways (i.e., the roadways of the intersection) rather than the roadway intersection itself. Such conventional techniques may therefore provide inaccurate risk assessments of roadway intersections, which, as will be appreciated, may cause a vehicle operator, fleet manager, etc. to unnecessarily plan a route through a particularly hazardous roadway intersection.

Thus, it is an objective of the present disclosure to provide advantageous systems and methods for predicting collision risk associated with roadway intersections. For example, in some embodiments, the systems and methods of the present disclosure may use one or more roadway intersection metrics in the generation of a collision probability for a roadway intersection. As will be described herein, such roadway intersection metrics may be particularly well-suited for providing accurate collision probabilities for roadway intersections. That is, the one or more roadway intersection metrics identified by the inventors of the present disclosure may have particularly useful predictive capabilities.

Additional advantages will be discussed below and will be readily apparent to those of ordinary skill in the art upon reading the present disclosure.

Reference will now be made in detail to example embodiments of the disclosure, wherein numerals refer to like components, examples of which are illustrated in the accompanying drawings that further show example embodiments, without limitation.

Referring now to, there is shown an example of a fleet management systemfor managing a plurality of assets equipped with a plurality of telematics devices. Each of the telematics devicesis capable of collecting various data from the vehicles(i.e., telematics data) and sharing the telematics data with the fleet management system. The fleet management systemmay be remotely located from the telematics devicesand the vehicles.

The vehiclesmay include any type of vehicle. For example, the vehiclesmay include motor vehicles such as cars, trucks (e.g., pickup trucks, heavy-duty trucks such as class-8 vehicles, etc.), motorcycles, industrial vehicles (e.g., buses), and the like. Each motor vehicle may be a gas, diesel, electric, hybrid, and/or alternative fuel vehicle. Further, the vehiclesmay include vehicles such as railed vehicles (e.g., trains, trams, and streetcars), watercraft (e.g., ships and recreational pleasure craft), aircraft (e.g., airplanes and helicopters), spacecraft, and the like. Each of the vehiclesmay be equipped with one of the telematics devices.

Further, it is noted that, while only three vehicleshaving three telematics devicesare shown in the illustrated example, it will be appreciated that there may be any number of vehiclesand telematics devices. For example, the fleet management systemmay manage hundreds, thousands, or even millions of vehiclesand telematics devices.

In some embodiments, the telematics devicesmay be standalone devices that are removably installed in the vehicles(e.g., aftermarket telematics devices). In other embodiments, the telematics devicesmay be integrated components of the vehicles(e.g., pre-installed by an OEM). As described herein, the telematics devicesmay collect various telematics data and share the telematics data with the fleet management system. The telematics data may include any information, parameters, attributes, characteristics, and/or features associated with the vehicles. For example, the vehicle data may include, but is not limited to, location data, speed data, acceleration data, fluid level data (e.g., oil, coolant, and washer fluid), energy data (e.g., battery and/or fuel level), engine data, brake data, transmission data, odometer data, vehicle identifying data, error/diagnostic data, tire pressure data, seatbelt data, airbag data, or a combination thereof. In some embodiments, the telematics data may include information relating to the telematics devicesand/or other devices associated with or connected to the telematics devices. Regardless, it should be appreciated the telematics data is a form of electronic data that requires a computer (e.g., a processor such as those described herein) to transmit, receive, interpret, process, and/or store.

Once received, the fleet management systemmay process the telematics data obtained from the telematics devicesto provide various analysis, predictions, reporting, etc. In some embodiments, the fleet management systemmay process the telematics data to provide additional information about the vehicles, such as, but not limited to, trip distances and times, idling times, harsh braking and driving, usage rates, fuel economy, and the like. Various data analytics may be implemented to process the telematics data. The telematics data may then be used to manage various aspects of the vehicles, such as route planning, vehicle maintenance, driver compliance, asset utilization, fuel management, etc., which, in turn, may improve productivity, efficiency, safety, and/or sustainability of the vehicles.

A plurality of computing devicesmay provide access to the fleet management systemto a plurality of users. The usersmay use computing devicesto access or retrieve various telematics data collected and/or processed by the fleet management systemto manage and track the vehicles. As will be appreciated, the computing devicesmay be any suitable computing devices. For example, the computing devicesmay be any type of computers such as, but not limited to, personal computers, portable computers, wearable computers, workstations, desktops, laptops, smartphones, tablets, smartwatches, personal digital assistants (PDAs), mobile devices, and the like. The computing devicesmay be remotely located from the fleet management system, telematic devices, and vehicles.

The fleet management system, telematics devices, and computing devicesmay communicate through a network. The networkmay comprise a plurality of networks and may be wireless, wired, or a combination thereof. As will be appreciated, the networkmay employ any suitable communication protocol and may use any suitable communication medium. For example, the networkmay comprise Wi-Fi™ networks, Ethernet networks, Bluetooth™ networks, near-field communication (NFC) networks, radio networks, cellular networks, and/or satellite networks. The networkmay be public, private, or a combination thereof. For example, the networkmay comprise local area networks (LANs), wide area networks (WANs), the internet, or a combination thereof. Of course, as will also be appreciated, the networkmay also facilitate communication with other devices and/or systems that are not shown.

Further, the fleet management systemmay be implemented using one or more computers. For example, the fleet management systemmay be implements using one or more computer servers. The servers may be distributed across a wide geographical area. In some embodiments, the fleet management systemmay be implemented using a cloud computing platform, such as Google Cloud Platform™ and Amazon Web Services™. In other embodiments, the fleet management systemmay be implemented using one or more dedicated computer servers. In a further embodiment, the fleet management systemmay be implemented using a combination of a cloud computing platform and one or more dedicated computer servers.